Youth as a Motor of Protest? What Polling can Tell us About the Young People who Demonstrate

There are two ways to answer the question “Who protests?”: One is to ask people about their personal motivationsexpand in interviews. Another is to use polling dataexpand to try to find out how people who go to protests differ from the population at large.

In April of 2018, Gwendolyn Sasseexpand and Félix Krawatzekexpand of the Centre for East European and International Studiesexpand in Berlin (ZOiS) conducted an online poll of 2000 young people from urban areas in Russia. The poll, among other things, asked about protests. Here, we consider the approximately 1,100 respondents who answered all the relevant questions.

Obviously, the purpose of this poll was not simply to find out about the respondents: the researchers hoped to apply the results to the people who were not surveyed directly (this is sometimes called “generalizing the findings”). However, this particular survey only permits conclusions to be drawn about people in the 16-34 age group who live in Moscow and St. Petersburg, the two biggest Russian cities, and other major urban centres, like Novosibirsk or Samara. This group is far from insignificant for the topic of protest, though, since young people are often a driving force for protest in big cities. This was true for the Bolotnaya movement in 2012, and again for the anti-corruption protests in 2017.

First, the simple question: What percentage of people took part in political, social or environmental protests in the past 12 months? To determine this, researchers combine the data from two questions – one asking about political and social protest activity, and one about environmental protest activity. The separate percentages for the two questions are shown in a chart further down.

This chart shows that protest activity is in no way a common phenomenon: only 8.5 percent of those surveyed had taken part in a protest within the previous year – i.e. less than one in tenexpand.

Bottom line: Young people are often the main drivers of protest, but this certainly does not mean that protests draw a majority of young people.

If one tries to sort out the different issues being addressed at protests, one finds that about four percent of respondents had taken part in environmental protests and about five percent had participated in political/social protests (the survey questions grouped these together).

This leads directly to the question of what characteristics people who protest have in common – and how they differ from those who prefer to stay home. Researchers who ask themselves this question tend to construct hypotheses – statements about a suspected link that can be confirmed or refuted through observation. These hypotheses can be drawn from scholarship or from the researchers’ own thoughts and observations. They are then tested using systematically acquired data.

As an example, we will claim in Hypothesis #1 that age plays a role in determining who attends social and political protests even within the 16-34 age group, insofar as younger people protest more than older people do – after all, the newspapers reported that “school children” were taking to the streets in 2017-18.

The line shows: The older a person is, the more unlikely it is that he or she has participated in a political or social protest within the last year.

Why “probability”? Because this study wants to make statements about the entire population of young people living in major urban centres – not just about those who actually participated in the survey. So, reading this graph we could say: if we pick 100 averageexpand sixteen-year-olds from the population of Russian big-city dwellers, we can assume that about six of them will have taken part in a protest within the last year. By contrast, only one out of a hundred average 34-year-olds will have done so.

Are there other factors that influence protest activity? Researchers examining geographic trends for protests in Russia report a concentration of protest activity in Russia’s “capitals” (stolitsy), Moscow and Saint Petersburg. Whereas in the years 1997-2000, only about one-tenth of the protests that took place in Russia were held in Moscow, in 2011, Moscow was the site of over 30 percent of protestsexpand. In absolute numbers Saint Petersburg ranges second behind Moscowexpand. Thus, it is possible that where a person lives has an effect on the probability that he or she participates in protest. This, then, is Hypothesis #2: people in Moscow and Saint Petersburg simply protest.

If this hypothesis is correct, then the red curve should lie above the black curve. This would indicate that residents of the two biggest cities have a higher probability of protesting, regardless of their age.

But: Fail!! (This, by the way, is a feeling that researchers experience often). The red curve does run above the black one, but the two are too close together to support an assertionexpand that there is a difference between the two groups.

In this case, the data fails to bear out the posited link. But what might at first look like a disappointment turns out to provide an important piece of information: The data shows that young urban people outside of the capitals are just as active (at least with respect to protest) as those living in the capitals. This makes it clear, once again, that there is no justification for limiting our focus only to Moscow.

Now the focus shifts to the economic situation of the respondents. Researchers who study political participation tend to assume that a person has to have a certain level of resources in order to participate in political activities. Simply put: a person who is working three jobs and still can’t make ends meet is not going to spend the little free time he has time going to demonstrations (see the collective goods problemexpand). Conversely, one could argue that people who are socially disadvantaged in particular have strong reasons to demand political and social changeexpand. So, one could phrase Hypothesis #3 to reflect either argument. Here is what the data have to say on the subject:

Another hypothesis failure: There is no systematic and significant difference among the groups. The researcher is starting to get nervous.

Incidentally, (no charts this time) neither level of education nor sex have a significant effect on the probability of protest activity. Okay, so what does then?

Is it possible that political protest has something to do with people’s trust in the president? This does seem likely. To find out, we use the data on how the survey participants used a four-part scale to describe their level of trust in Putin (naturally, controlling for all other factors). Hypothesis #4 says: The less trust people have in Putin, the higher the probability that they engage in protest.


The key aspect here: the probability does not change gradually with the level of trust. Only the group that reported not trusting Putin “at all” is clearly separated from the other three. Among the other three groups, there are no significant gradations. This indication of a polarisation of protest behaviour suggests that a lesser degree of trust is not associated with a greater willingness to protest – until, that is, the level of trust falls below a certain threshold. Since people who do not trust Putin “at all”, at just under 14%, make up the smallest group in the survey population,

we can now see why the overall percentage of protesters is not particularly high.

Hence political/social protest is clearly associated with an anti-Putin stance. But what happens when we look at environmental protest …

… once again, we see that the respondents who trust Putin the least are most likely to have engaged in protest. But since the confidence intervals overlap, there are insufficient grounds to assert that this difference exists in the population as a whole. Environmental protest then – at least on the basis of these data – cannot by any means be described as always having to doexpand with Putin.

We are not done yet, though, because no scientific study is complete without the Conclusion section.

So here it is.

  1. The young people living in big cities who have engaged in protest activity of any kind constitute a small minority.
  2. Political protest activity appears to have something to do with age, including for the 16-34 age group studied here: younger people have a significantly higher probability of protest activity.
  3. Protesters as a group do not share any of the sociodemographic characteristics examined in this study other than age: no statistical effects are associated with education, living standard, sex, or city of residence.
  4. There is – no real surprise here – a link between the level of trust in Putin and the probability of protest activity: the lower the former, the higher the latter.
  5. This link – like the age effect – can be observed in the case of political and social protest but vanishes when one looks at environmental protest.

Just one little disclaimer: All of these results could very easily depend on when the survey was conducted. After all, Navalny mobilised a great many young people to take part in political protest in 2017 and 2018expand. It’s possible that things will look very different a year later.

Jan Matti Dollbaum studied political science and Slavonic studies in Heidelberg, St. Petersburg, Mainz and London. Since summer 2016 he has been a PhD student at the Research Centre for East European Studies at the University of Bremen. His dissertation, which is currently in its final phase, deals with the political and social conditions of protest development after large waves of mobilization – with a focus on local protest in Russia.